KBB-Forum 2003 , Cilt 2 , Sayı 2


Sinan KOCATÜRK, MD1; Tayfur DEMİRAY, MD2; Tanzer ÇAKIR, MD1; Gül BAHAR, MD2; Ünsal ERKAM, MD1; Ali MERT, MD2
1SSK Ankara Eğitim Hastanesi, II. KBB Kliniği, Ankara, Türkiye
2SSK Ankara Eğitim Hastanesi, Mikrobiyoloji Kliniği, Ankara, Türkiye


AIM: To determine the transient bacteriemia ratios during the elective tonsillectomy and compare the bacteria detected with superficial/central tonsillar cultures.

METHODS: 46 patients were included in this study. Preoperative surface swab cultures, intraoperative tonsil central swab cultures, preoperative and postoperative blood culture samples were obtained. Antibiotic sensitivity tests were done for the detected pathogen bacteria.

RESULTS: Bacteriemia was detected in six patients (%13). In 5 of the patients (%83.3), isolated microorganism (S.aureus) in the blood culture was the same as the one in central swab cultures. In 5 of the 6 cases with bacteriemia (%83.3) resistance to penicillin was detected.

CONCLUSION: Bacteriemia at the rate of 13% and resistance to penicillin at the rate of 83.3% warrants the necessity of antibiotic prophylaxis especially in risky patients. The similarity of the pathogens detected in the blood to the central tonsil pathogens at the rate of %83.3 (5/6) suggested that it was not appropriate to choose an antibiotic based on superficial tonsil cultures.


While transient bacteriemia due to tonsillectomy does not cause any problems in healthy individuals, it may cause high mortality in the risk group with congenital/acquired heart disease or orthopedic prosthesis despite antibiotic treatment [1,2]. To counter this probability antibiotic prophylaxis is being frequently administered in risky patients [3,4].

Bacteriemia observed during tonsillectomy may develop due to microorganisms in the central region of the tonsil or contaminated oropharyngeal secretions or due to local infections [5]. It is known that tonsil surface cultures do not reflect central tonsil cultures [5]. Therefore, it may be wrong to make a decision about prophylactic antibiotic choice solely based on surface culture results. The identification of microorganisms observed during bacteriemia is significant in choosing an antibiotic especially for risky patients.

This study has been conducted with the aims of determining the transient bacteriemia ratios in tonsillectomy cases that underwent classical dissection and comparing the bacteria detected in bacteriemia with surface/central tonsil cultures.


46 patients who underwent elective tonsillectomy with the diagnosis of chronic recurrent tonsillitis in our clinic between April 2002 and September 2002 were included in the study. 34 of the patients were male and 12 were female and their average age was 5.4. The indications for tonsillectomy were: not less than 5 attacks of tonsillitis per year, persisting for at least 2 years; high fever, pain, tonsillar hypertrophy and hyperemia, cervical lymphadenopathy and persisting complaints despite recurring antibiotic treatment.

Care was taken that the patients included in the study did not have any attacks of tonsillitis or upper respiratory tract infections 4 weeks prior to the operation and did not use antibiotics for any reason for at least 20 days before the operation. In the postoperative period oral amoxicillin clavulanic acid suspension was used for 5 days for prophylaxis.

Immediately after the induction of anesthesia, preoperative blood culture was taken for control purposes in accordance with the techniques of sterile blood collection from peripheral vein. Immediately before the operation, taking pains not to cause oropharyngeal contamination swab cultures were taken from both tonsils and planted beside the operation table. The tonsils which were removed with the dissection technique were kept in povidin-iodine (Batticon) solution for 35-45 seconds to prevent contamination from outer portions of the tonsils and then bathed in sterile physiological saline. Then after the tonsils were separated into two with the help of a sterile lancet, swab cultures were taken from the central regions with sterile cotton swabs. Immediately after the removal of the tonsils (in 2 minutes) blood culture was taken from the peripheral vein. The sample was plated onto aerob and anaerob blood culture media (BACTEC 9050; Becton, Dickinson and Company, Franklin Lakes, NJ). Surface swab and central swab culture samples were plated onto 5% bovine blood growing medium, chocolate agar and eosin-methylen blue agar (EMB) growing mediums. After incubation in a 5% CO2 containing environment at 35° C for 48 hours, the growths were assessed using standart microbiological methods. Microorganisms accepted as possibly pathogen that grow dominantly beside the polymicrobial normal throat flora in surface and central swab cultures were tested for antibiotic sensitivity in the Mueller-Hinton growing medium by the Kirby-Bauer disc-diffusion method.


No growth was identified in the preoperative blood culture samples. In postoperative blood culture samples bacteriemia was detected in 6 patients (13%). In five of the patients (83.3%), it was observed that the possible pathogen microorganism isolated in the blood culture was the same as the one in the central swab cultures. In one patient; although no pathogens were identified in surface and central cultures, bacteria that could be pathogen grew in blood culture (Table 1). No anaerobic bacteria were identified in postoperative blood cultures.

Of the 6 patients in whom bacteriemia was detected, in 4 patients S.aureus and in 1 patient group A beta hemolytic streptococcus (GSBHS) and Haemophilus spp. were cultivated. It was observed that pathogens growing in blood cultures were not compatible with surface cultures (Table 1).

Microorganisms that could be pathogens (presumptive pathogens) were identified in 16 patients in surface swab cultures (34.78%) and in 27 patients in tonsil central swab cultures (58.69%). The most frequently encountered pathogens in surface cultures were respectively S.aureus, GABHS (group A beta hemolytic streptococcus=Streptococcus pyogenes) and Haemophilus spp. In tonsil central swab cultures, on the other hand, S.aureus, Haemophilus spp and GABHS were identified according to order of frequency (Table 1).

In superficial and central cultures, while the same pathogen was identified in 9 patients (19.56%), in 18 patients (39.13%) different pathogens were identified (Table 1). According to the antibiotic sensitivity test results; all bacteria except GABHS were identified as penicillin-resistant. The antibiotic sensitivities of bacteria isolated in blood cultures are shown in Table 2.

Table 1: Isolated bacteria
NBF: Normal throat flora,
MSSA: Methycilline sensitive S.aureus
MRSA: Methycilline resistant S.aureus
GABHS: Group A beta hemolytic streptococcus

Table 2: Antibiotic sensitivity test results
MSSA: Methycilline sensitive S.aureus
MRSA: Methycilline resistant S.aureus
GABHS: Group A beta hemolytic streptococcus

Including the ones having a positive blood culture none of our patients developed infective complications.


Theoretically, there is a risk of bacteriemia due to operation in regions where there is bacterial flora. Yet, bacteriemia ratios are differential depending upon the site of operation. In the septoplasty operation, although there is normally S.aureus colonization in the nasal mucosa, the risk of bacteriemia is very little [6]. Similarly, although there is a theoretical risk in cases with ventilation tube insertion, bacteriemia ratios have not been completely verified [6,7].

Transient bacteriemia may lead to serious complications in the risky patient group while causing no problems in healthy patients [5]. The effect of transient bacteriemia due to tonsillectomy especially on the development of endocarditis in patients with cardiovascular risks is very well known [8]. Therefore; there is a consensus on the use of prophylactic antibiotics especially in the risky patient group. It has also been reported that prophylactic antibiotic treatment reduces bleeding and postoperative pain and increases recovery [8,10].

Transient bacteriemia may develop as a result of bacterial diffusion through the veins in the tonsillary tissue or the pharyngeal mucosa and through the open wound margins [4]. The traction applied before starting the dissection may also have a role in bacterial diffusion [9]. The findings that in 5 of the 6 patients in whom we identified bacteriemia the bacteria growing in the blood were the same as the bacteria growing in the central tonsil culture and that the bacteria growing in superficial cultures were not identified in blood cultures have suggested that transient bacteriemia may originate from tonsil central bacteria. The traction and squeezing of the tonsil may have been effective in this result.

While post tonsillectomy transient bacteriemia ratios were given as 22%[13], 25%[12], 41%[8] in the literature, in our series this ratio was found as 13% (6/46). The differences among bacteriemia ratios may be attributed to different blood culture methods and blood culture collection times [2]. It has been reported that transient bacteriemia occurs within a one-hour time period [14]. There are different approaches regarding the timing of culture collection such as: immediately after removing the first or second tonsil [15] , within the first 5 minutes after tonsillectomy [16], immediately after the completion of the operation [17], 2 minutes after the removal of the second tonsil [2,18], during tonsillectomy [5], in the postoperative period [12]. In our study the blood culture was taken within 2 minutes following the completion of the tonsillectomy operation.

Post-tonsillectomy bacteriemia ratios may also depend on the surgical technique. Gaffney et al. reported that bacteriemia ratios were lower in tonsillectomies performed with the guillotine method compared to those performed with the dissection method and that this could be due to the guillotine’s compression on the tonsillar blood vessels [18]. Conversely, Olina et al. reported a 60% rate of bacteriemia with the guillotine technique while detecting only a 19% rate of bacteriemia with the dissection technique [19]. Walsh et al., on the other hand, found no statistically significant differences between the two techniques with respect to the incidence of bacteriemia [2].

Streptococcus pyogenes (GABHS) is stated to be responsible in the etiology of endocarditis, arteritis, and osteomyelitis and it is also reported to be capable of leading to serious mortality in patients with cardiovascular risk factors despite antibiotic treatment [20]. In the series of Rhoads et al. comprised of 68 patients, Streptococcus pyogenes (GABHS) was cultivated in the blood cultures of 4 patients [17]. In our series we identified Streptococcus pyogenes in one patient (1/46).

Another microorganism responsible in the aetiology of endocarditis is the alpha hemolytic streptococcus [21]. Kaygusuz et al. identified alpha hemolytic streptococcus in one case [12]. In this series no alpha hemolytic streptococci were identified.

H. İnfluenza serotype b may cause invasive bacterial infections in children under 3 years of age[22,23]. In our case series, in central swab cultures Haemophilus spp. were cultivated in 7 (7/46) cases. In postoperative blood cultures on the other hand Haemophilus spp. was identified in one out 6 patients in whom growth was identified.

In the 32 case series of Francois et al. anaerobe bacteria were not cultivated in blood cultures [5]. Kaygusuz et al. reported anaerob bacteria growth in one case.12 In our case series there was no anaerobic growth.

S.aureus may lead to serious systemic infections such as pneumonia, osteomyelitis, acute endocarditis, pericarditis, meningitis through bacteriemia besides causing local infections [24]. It is worth noticing that in our case series we identified S.aureus in 4 of the 6 patients detected to have bacteriemia (4/6) and that these showed similarities to the central cultures. 3 out of 4 S.aureus cultivated samples were of the type sensitive to methycilline (MSSA) and 1 was of the type resistant to methycilline (MRSA). With respect to their sensitivities to methycilline, detection of similarities with central swab cultures supports the idea that bacteriemia originates from tonsil central bacteria.

None of our patients, including those with a positive blood culture, developed any infective complications. This finding suggested that the number of bacteria seen in the blood during bacteriemia was below 10CFUs/mL; therefore, it can be stated that risk of metastatic infection is extremely low in healthy children [25].


The growth of the same pathogen bacteria in blood and central swab cultures in 5 patients detected to have bacteriemia suggested that the bacteriemia could originate from tonsil central bacteria. Our detection of bacteriemia at the rate of 13% revealed the necessity of antibiotic prophylaxis especially in risky patients. In order to determine the antibiotic of choice in prophylaxis we think that studies adopting larger case series which reflect the microbiological profile of the Turkish people are warranted.


1) Carithers JS, Gebhart DE, Williams JA. Postoperative risk of pediatric tonsilloadenoidectomy, Laryngoscope 1987;97(4): 422-429 , [ Özet ]

2) Walsh, R.M, Kumar B.N, Tse A, Jones P.W, Wilson P.S. Post-tonsillectomy bacteremia in children. J Laryngol Otol. 1997;111(10): 950-952., [ Özet ]

3) Durack DT. Current issues in prevention of infective endocarditis. Am J Med. 1985;28;78(6B): 149-156., [ Özet ]

4) King RC, Crawford JJ, Small EW. Bacteremia following intraoral suture removal. Oral Surg Oral Med Oral Pathol. 1988;65(1): 23-28. , [ Özet ]

5) Francois M, Bingen EH, Lambert-Zechovsky NY, Kurkdjian PM, Nottet JB, PNarcy P. Bacteremia during tonsillectomy. Arch Otolaryngol Head Neck Surg. 1992;118(11): 1229-1231. [ Özet ]

6) Silk KL, Ali MB, Cohen BJ, Summersgill JT, Raff MJ Absence of bacteremia during nasal septoplasty. Arch Otolaryngol Head Neck Surg. 1991;117(1): 54-55.

7) Lohr JA, Sloop FB, Sydnor A, Donowitz L. Bacteremia associated with tympanostomy tube insertion. J Infect Dis. 1989;159(3): 594-595 , [ Özet ]

8) Abrahamson L, Subacute bacterial endocarditis following removal of septic foci. British Medical Journal 1931;2 : 8-9

9) Soldado L, Esteban F, Delgado-Rodriguez M, Solanellas J, Florez C, Martin E. Bacteremia during tonsillectomy : a study of the factors involved and clinical implications Clin Otolaryngol. 1998;23(1): 63-66., [ Özet ]

10) Telian SA, Handler SD, Fleisher GR, Baranak CC, Wetmore RF, Potsic WP. The effect of antibiotic therapy on recovery after tonsillectomy in children. A controlled study. Arch Otolaryngol Head Neck Surg. 1986;112(6): 610-615., [ Özet ]

11) Jones J, Handler SD, Guttenplan M, Potsic W, Wetmore R, Tom LW. The efficacy of cefaclor vs amoxicillin on recovery after tonsillectomy in children. Arch Otolaryngol Head Neck Surg. 1990;116(5): 590-593., [ Özet ]

12) Kaygusuz I, Gök U, Keleş E, Kizirgil A, Demirbağ E. Bacteremia during tonsillectomy. Int J Pediatr Otorhinolaryngol. 2001;58(1): 69-73 , [ Özet ]

13) Anand VT, Phillipps JJ, Allen D, Joynson DH, Fielder HM. A study of postoperative fever following paediatric tonsillectomy. Clin Otolaryngol 1999;24(4):360-364 , [ Özet ]

14) Van Eyck M. Bacteremia after tonsillectomy and adenoidectomy. Acta Otolaryngol 1976;81(3-4): 242-243 , [ Özet ]

15) Rubin M.I, Ebstein I.M, Werner M. Blood cultures after tonsillectomy. Am J Dis Childhood 1929;38: 726-729,

16) Elliot S.D. Bacteremia following tonsillectomy. Lancet 1939;2: 589-592

17) Rhoads PS, Sibley JR, Billings CE, Bacteremia following tonsillectomy. Effect of preoperative treatment with antibiotics in postoperative bacteremia and in bacterial content of tonsils, J.Am.Med Assoc. 1955;157: 877-881

18) Gaffney R.J, Walsh M.A, McShane D.P,Cafferkey M.A. Post tonsillectomy bacteremia. Clin Otolaryngol. 1992;17(3): 208-210., [ Özet ]

19) Olina M, Garavelli PL, Grosso E, Gugliemetti C, Pia F. Bacteremia in tonsillectomy: Sluder’s technique versus dissection. Preliminary results. Recenti Prog Med 2001;92:2:121 Recenti Prog Med. 2001;92(2):121-124, [ Özet ]

20) Wong VK, Wright HT Jr. Group A beta hemolytic streptococci as a cause of bacteremia in children Am J Dis Child. 1988;142(8): 831-833., [ Özet ]

21) Feldman L, Trace IM. Subacute bacterial endocarditis following removal of teeth and tonsils, Ann. Intern. Med. 1938;11: 2124-2132

22) Friedman E, Damion J, Healy G, McGill TJ. Supraglottitis and concurrent Haemophilus meningitis. Ann Otol Rhinol Laryngol. 1985;94(5 Pt 1): 470-472., [ Özet ]

23) Koo W, Oley C, Munro R, Tomlinson P. Systemic Haemophilus influenzae infection in childhood. Med J Aust. 1982;2: 77-80, [ Özet ]

24) Kloos WE, Bannerman TA. Staphylococcus and micrococcus. In: Murray PR, Baron EJ, Pfaller MA, Tenover FC, Yolken RH (eds). Manual of Clinical Microbiology. 6th ed. Washington: ASM Press, 1995.p. 282-298

25) Sullivan TD, LaScolea LJ, Neter E. Relationship between the magnitude of bacteremia in children and the clinical disease. 1982; 69:699-702 Pediatrics. 1982; 69(6): 699-702. [ Özet ]